Computational Modeling of Diphosphine Oxide and Diglycolamide Ligand Complexation to Lanthanides and Extraction from Acidic Media

Certain "critical materials" such as lanthanides have sparked a research interest in the greater scientific community; problems associated with lanthanide acquisition, such as environmental hazard that is caused by the generation of highly acidic or basic waste byproducts, have inspired studies into novel methods of separation of these materials from the ores in which they reside. While computational methods can be used to explore such problems in detail, an adequate theoretical representation of the system under consideration must first be devised. In this study, three different ligand-lanthanide- organic solvent extraction systems were analyzed, using a variety of different computational models to represent each extraction system. In each case, the Δ(ΔG) value of the complexation reaction was computed for a variety of organic media, with the end goal of finding a computational model that has a maximal coincidence with experimental results for each extraction system. For diphosphine oxide extraction of lanthanides from nitric acid, the ΔGrxn trends show proper correlation with experimental logD values when the polarizable continuum model (PCM) and the conductorlike screening model (COSMO) are used as implicit solvation models, but not when the self-consistent reaction field (SCRF) implicit solvation model is used. For the diglycolamide ligand extraction of lanthanides from hydrochloric acid, the two computational models provide qualitative accuracy, when modeling systems with a 1:1 ligand-to-metal ratio but break down when systems with a higher ligand-to-metal ratio are considered. Finally, for the diglycolamide ligand extraction of lanthanides from nitric acid, the opposite ΔG trend is observed compared to experimental logD values.